Following is project structure:
MVC 4 application
1) Project is for MVC having controller has injected interfaces.
2) interfaces are existed into this application.
Manager library project.
1) Project contain refernece of MVC application.
2) Implementation of each interfaces into this assembly.
3) NinjectMOdule is existed into this project where binded interface with related manager class.
PROBLEM: unable to get injected classes into web application.
it gives following error:
Error No matching bindings are available, and the type is not self-bindable.
Activation path:
2) Injection of dependency ILeadInformation into parameter leadInformation of constructor of type HomeController
1) Request for HomeController
Usually in the "root" project which manages the kernel you need to load all the modules. This has to be done during initialization of the application i.E. before the application relies on the binding being there.
Let's say the module which binds the ILeadInformation is called LeadModule. It would look like:
IKernel.Load<LeadModule>();
EDIT:
For your concrete scenario, you have your MVCApplication Assembly which contains the application and some interfaces and you have your Manager assembly, which references the MVCApplication assembly, like so:
Now you ought to place the Composition Root in the MVCApplication assembly. But, uh no, you can't reference the Manager assembly, since that would lead to a circular dependency.
So we got to fix this. We need to move the interfaces into a new assembly. And adjust the references so it will look like:
Related
I am getting System.IO.FileNotFoundException in my .Net Core Web API. So I've set up the below project to demonstrate the problem.
I created a.Net Standard library named DemoLibrary and added QRCoder dependency via NuGet.
Disclaimer: Reason for choosing the QRCoder is that the Web API doesn't use it by default. I don't use it in my project. In fact, I'm getting this exception for EntityFrameworkCore.
I created a new .Net Core Web API DemoWebAPI which has no other dependencies.
Then added the DemoLibrary to DemoWebAPI via Add Reference -> Browse -> DemoLibrary.dll.
This is my solution:
The DemoMethod method in Calculate class just creates the object of QRCodeGenerator.
public class Calculate
{
public static string DemoMethod()
{
QRCodeGenerator qrGenerator = new QRCodeGenerator();
return "";
}
}
And my ValuesController in DemoWebAPI just calls the method:
[HttpGet]
public ActionResult<IEnumerable<string>> Get()
{
return new string[] { "value1", "value2", DemoLibrary.Calculate.DemoMethod() };
}
Now, when I run the DemoWebAPI project I get below exception upon the call to the DemoMethod:
System.IO.FileNotFoundException: 'Could not load file or assembly 'QRCoder, Version=1.3.5.0, Culture=neutral, PublicKeyToken=null'. The system cannot find the file specified.'
I understand the fact that I have to copy the QRCoder.dll file somewhere. But I fail to understand where to put it. I've already tried putting it in "bin/debug/netcoreapp2.2" of the DemoWebAPI and "bin/debug/netstandard2.0" of the DemoLibrary.
But I couldn't get it working.
Request: Please post your answer as descriptive as you can because I am new to .Net Core.
Edit:
I am aware of the NuGet servers. I have read topics like hosting a NuGet server in IIS and Azure. The reason behind DLL reference is I want to use my DLLs in two projects one of them is a .net core API and the other is .net framework class library which is compiled by NMAKE. I couldn't find any way to restore NuGet packages in the .MAK files.
It looks like you've merely added the DLL for DemoLibrary to your DemoWebApi project. That's not how you should be adding references. Since these are in the same solution, you should add a project reference. That will fix your issue.
Now, let me explain what's actually going on here. Your DemoLibrary has a dependency on QRCoder. It's a NuGet reference, which means that package will be restored (i.e. downloaded) and included in your DemoLibrary build output. However, it will be included as one or more DLLs along side your DemoLibrary.dll. When you then just reference DemoLibrary.dll, you're missing all these other DLLs that are part of DemoLibrary and thus, things don't work properly.
Now, when it comes to a project reference, things are little more complex. A project reference essentially wraps the referenced project into your other project. You can think of it as sort of a sub project. For all intents and purposes, it's like any dependency of the sub project becomes a dependency of the main project. That means that DemoWebAPI now technically has a NuGet package reference to QRCoder even though there's no explicit package reference in its project file. The dependency comes from your DemoLibary project. As such, with a project reference, all the necessary dependencies will be included, because it's as if the main project included those itself, by way of the sub project.
For what it's worth, you should virtually never include a DLL as a reference directly. That used to be required, but the concept of NuGet packages has all but eliminated the practice. Even if DemoLibrary was not in the same solution as DemoWebAPI (meaning you could no longer do a project reference), the correct way to use it would be to turn DemoLibary into a NuGet package, and then reference it in DemoWebAPI via a package reference, like any other NuGet package. You do not simply add the DLL.
I'm creating a class library with integration tests. I'm using combination of FsCheck and xUnit here. Tests are supposed to run queries on the database to verify an output. I'm using FSharp.Data.TypeProviders with System.Data.Linq (so Linq2Sql type provider) to access the database - type provider is defined inside test class library itself.
However when I'm trying to build a project, I'm getting a Static linking may not use assembly that targets different profile. error. Class library, where tests are located is targeting .NET Framework 4.5 (FSharp runtime 4.4.0). I've also set binding redirects for FSharp.Core assembly, however this doesn't change anything.
The Startup class in asp.net 5 strikes me as an odd duck. It isn't class Startup : ISomething or Startup : BaseSomething where the interface or base class is part of some Microsoft.AspNet.* assembly. No Startup is just a plain class with the correct magical method signatures created by convention.
How is execution passed from DNX to Startup.ConfigureServices?
Lets take for example calling:
dnx.exe . web
So the . tells dnx that it can find the project.json in the current folder. From there is finds the command associated with the key "web". So if the local project.json has this:
"commands": {
"web" : Microsoft.AspNet.Hosting --server Microsoft.AspNet.Server.WebListener --server.urls http://localhost:5000"
}
I am going to take a stab that combined it would be the equivalent of:
dnx.exe . Microsoft.AspNet.Hosting --server Microsoft.AspNet.Server.WebListener --server.urls http://localhost:5000"
I also get how the dnx gathers all the sources (including dependencies) using parameters in the project.json for in memory compilation so we now have the user's assembly "MyAssembly" and all dependent assemblies available to the dnx. The dnx has loaded Microsoft.AspNet.Hosting as a managed entry point. So execution passes from the unmanaged "stub" to Microsoft.AspNet.Hosting managed assembly. Correct so far?
The next parameters are instructing Microsoft.AspNet.Hosting that it will be hosting an instance of Microsoft.AspNet.Server.WebListener (specifically on port 500 of localhost). Ok so how does Microsoft.AspNet.Server.WebListener "know" to look for a class named specifically "Startup" in "MyAssembly". Is it simply hard coded into Microsoft.AspNet.Server.WebListener? Into Microsoft.AspNet.Hosting?
The jump to Startup class seems to be the last "magic". Execution before and after that is starting to get pretty clear but I feel I am still missing something.
DNX knows how to load and execute assemblies that have class named Program which has a method named Main. You are passing Microsoft.AspNet.Hosting as the startup assembly to DNX when you run the web command.
Hosting has a Main method.
This code that eventually gets called from the Main method mentioned above has "Startup" hardcoded in it. The actual hardcode is here.
I'd like to add some details to Victor's answer:
DNX's logic that finds the entry points supports both static and instance Main methods.
There's no coupling between the server (WebListener on this case) and the Startup class nor RequestDelegate. The glue if made via a callback supplied by the hosting layer to the server factory.
The transition is not direct from unmanaged code to ASP.NET hosting. There's a managed entry point/application host on DNX itself. This is where the support for the new project structure is implemented (still independent from ASP.NET).
I wrote a series of posts that dives into the details of bootstrapping an ASP.NET 5 application, from the DNX native host up to request handling. All those execution details are covered on the posts.
I have an .ear file with the standard lib directory.
I have a .jar file in that lib directory. It contains UserInfoManager, which is an interface. It contains (for these purposes) no other classes. It also contains a META-INF/beans.xml file.
I have another .jar file in that lib directory. It contains a class named UserInfoResource that is a JAX-RS resource class. That class has the following inside it:
#Inject
private UserInfoManager userManager;
Next, I have an EJB .jar file at the root of the .ear file. It contains a class named UserManagerBean that implements the UserInfoManager interface. This class is annotated with #Stateless and basically nothing else (thus making it a local stateless session bean exposed via its local business interface (UserInfoManager). This .jar file also has a META-INF/beans.xml file.
Next, I have a .war file with an Application class in it and nothing else. This serves as the "mounting point" for any and all JAX-RS resources discovered at deployment time present in the lib directory. I do not declare this Java EE 6 module as a CDI bean archive since it contains no beans.
This spec-compliant arrangement fails at deployment time. Weld (the CDI implementation in GlassFish 3.1.2) claims that the injection point detailed above cannot be satisfied, as there are no known implementations of UserInfoManager available to it.
When that injection point is annotated with #EJB instead, everything works fine.
How do I get CDI to inject a local stateless session bean reference into a JAX-RS resource that is present on the classpath?
Update: Because no matter how I look at this it seems like a specification violation, I have filed a bug with a testcase attached. I encourage readers to take a look and see if they can get it to work.
Update: The workaround is to make sure that your JAX-RS classes are not bean archives, but are annotated with #ManagedBean. In addition, the {{.war}} file that serves as their mount point must be a bean archive (must have a {{WEB-INF/beans.xml}} file). Some combination of these requirements is a CDI specification violation. The following bug tracks these issues: http://java.net/jira/browse/GLASSFISH-18793
Jersey does not treat Resources as managed beans unless there is an explicit scope/#ManagedBean annotation attached to it. So, you need to annotate your resource with #ManagedBean or #RequestScoped for the injection to work.
Seems the problem occurs only if beans.xml is included in the resource jar file. When I remove it and attach #ManagedBean annotation to the resource class (instead of #RequestScoped, since #RequestScoped does not work if no beans.xml is present) it works. I am not a CDI expert, so not sure if this is as designed or a bug.
This call
// this._cfg is an NHibernate Configuration instance
this._sessionFactory = this._cfg.BuildSessionFactory();
Gives me this exception at runtime (NOT at compile time).
Could not load file or assembly 'NHibernate.ByteCode.Castle' or one of its dependencies. The system cannot find the file specified.":"NHibernate.ByteCode.Castle
OK so far. But the thing is, this code is running in a class library project, and I have referenced NHibernate.ByteCode.Castle (along with all the other NHibernate dll's) in that project.
Wierder: I can fix the exception by additionally referencing the NHibernate dll's in the Windows WPF executable project that calls my class library. But the Windows WPF executable contains no code that directly uses NHibernate (as evidenced by: It compiles fine without any NHibernate references). So what's going on? Apparently it's insufficient to reference NHibernate.ByteCode.Castle in the project that actually uses the NHibernate stuff. Anyone know why?
I know this is old, but what I've done to fix the dependency problem is simple:
In my UnitOfWork I added one static method:
private static void bringCastleDamnit()
{
var pf = new NHibernate.ByteCode.Castle.ProxyFactoryFactory();
}
Then, and only then, would MSBuild see that it was needed and copy it to my output directory for my (asp.net and console) apps that references my Data project.
I wouldn't reference the castle byte code factory at all; just ensure it (and all other needed dependancies) are copied to the output directory using a post-build step.